Abstract
A fast accurate multi-stage numerical method, back-averaging, is developed, analyzed, and optimized. The technique is used to solve an alternating dimensional numerical simulation of fusion plasma transport based on nonlinear resistive magnetohydrodynamics (MHD) equations. The geometry of the model is a complicated time-dependent two-dimensional configuration of flux contours analogous to a doublet. Numerical convergence rate comparisons of the optimized back-averaging method with various other iterative techniques for solving the numerical problem show that optimized back-averaging is the fastest method of all those considered. Moreover, to accomplish convergence in a practical length of time for extremely peaked profiles and complicated time varying configurations, back-averaging is essential. Further, with a miminum of additional computation, optimized back-averaging yields extreme accuracy.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call